Re: Moving forward with the CCAMP charter

["Adrian Farrel" <adrian@olddog.co.uk>]

Hi Lyndon,

> As with everyone else, I appreciate all of the work that
> you put in to create a very detailed workplan for the group.

Thanks.

> Some follow up questions and suggestions:
>
> -- On the "aligning GMPLS across standards bodies" item:
>
> First of all, what standards bodies do you see included under
> this effort?

Any and every that is or plans to use GMPLS or derive a variant of GMPLS.

> Also, are you intending this to be a unilateral document on ccamp's
> part?

Yes.

> If it is, I'm not sure I see what it would say besides "use the RFCs or
> bring the requirements back to us", which is what this group has said in
> the past.  If on the other hand you see this as the basis for joint
> discussion with other bodies, it might be a very helpful activity.

First, I refuse to be hamstrung by history.
Note that alignment requires that we start from where we are now (with
non-interoperable variations of the protocols) and try to resolve the
problem.

Second, doing this work does not preclude discussing with other bodies.
In fact, the draft is likely to lead specifically to discussions with
other bodies.
However, it is not pragmatic or wise to discuss with other bodies the
actions which CCAMP thinks it should take. These are CCAMP actions.

> -- On the ""routing and signaling for complex optical constraints" item:
>
> The ashwood draft seems to me to have two components, one being
> requirements and considerations of routing across a purely photonic
> domain and the second being a particular solution (the matrix
> representation), to which there are alternatives such as the virtual
link
> approach that Igor has identified.
>
> I think these two should probably be separated, as the solution may be
> more widely applied to any situation of advertising a virtualized
> topology, such as in enhanced l1vpn.

You'll notice that the work item is stated as "Analysis and protocol
changes for routing and signaling for link viability constraints" whether
this turns into one, two or a hundred I-Ds cannot possibly be determined
at this point. I suspect that the issue of routing across a photonic
domain as described in the Ashwood I-D is only a special case of a generic
multi-domain TE issue. Hence this item appears under the Inter-Domain
heading and is described much more generally.

You are correct that some of this would feed into (overlap with) the L1VPN
enhanced mode.

Thanks,
Adrian